The first algorithm to play FlappyBird.
https://mp.weixin.qq.com/s/nmqX05c2qS3Jeg66W5NOzA
OS: Windows10
Python: Python3.5+(have installed necessary dependencies)
usage: flappybird.py [-h] [--mode MODE] [--policy POLICY]
Use q learning to play flappybird
optional arguments:
-h, --help show this help message and exit
--mode MODE Choose <train> or <test> please
--policy POLICY Choose <plain> or <greedy> please